Table Meta

It's a universal class that created to be a middleware, universal mapping for data from different parsers - simple-ddl-parser and py-models-parser.

Based on this middleware 2 libraries are worked - omymodels & fakeme.

It's allow create 1 adapter for different inputs and produce output only on one standard - easy to maintain ad add different output variants.

All classes - Pydantic classes, so you can do with them anything that you can with Pydantic classes.

Library contains 2 different classes - TableMeta - main class to convert input relative to models or tables. Second - Type, for Enum types data.

How it use

Install

    pip install table-meta

Usage

from table_meta import TableMeta

data = {your_table_definition}

table_data = TableMeta(**data)

Convert simple-ddl-parser input to TableMeta

Simple-ddl-parser: https://github.com/xnuinside/simple-ddl-parser

Pay attention that TableMeta expected data from simple-ddl-parser , that created with flag 'group_by_type=True' Example: result = DDLParser(ddl).run(group_by_type=True, output_mode="bigquery")

To convert simple-ddl-parser output to TableMeta - use method: ddl_to_meta()

Usage example:

    from simple_ddl_parser import DDLParser
    from table_meta import ddl_to_meta

    ddl = "your ddl"
    parser_result = DDLParser(ddl).run(group_by_type=True, output_mode="bigquery")
    data = ddl_to_meta(parser_result)

    # ddl_to_meta returns Dict with 2 keys {"tables": [], "types": []} inside lists you will have Table Meta a models

    print(data)

Convert py-model-parser input to TableMeta

Py-models-parser: https://github.com/xnuinside/py-models-parser

Usage example:

    from py_models_parser import parse
    from table_meta import models_to_meta

    model_from = "your python models, supported by parser"
    result = parse(model_from)
    data = models_to_meta(result)

    # models_to_meta returns Dict with 2 keys {"tables": [], "types": []} inside lists you will have a Table Meta models

    print(data)

Changelog

v0.3.2

1. Added 'comment' attr to the Column

v0.3.0

1. Added search for type in columns with foreign keys from models
2. Fixed issue with popylating Column details like unique or primary key from py-models-parser

v0.2.2

1. Added if_not_exists to table properties

v0.2.1

1. Added support for parsing 'dataset' from data as 'table_schema' also added fields like 'project' (to support BigQuery metadata)
2. Depencencies updated
3. Added HQL Table Properties

v0.1.5

1. field 'attrs' added to Type to store values from py-models-parser output

v0.1.3

1. 'parents' added to Type and to Table

v0.1.1

1. Fix dependencies for python 3.6

v0.1.0

1. Table Meta moved from O!MyModels to separate library. To make it re-usebale in fakeme library.